BACKGROUND

Angiogenesis can be enhanced by several growth factors, like vascular endothelial growth factor-165 (VEGF(165)) and basic fibroblast growth factor (bFGF). Delayed release of such growth factors could be provided by incorporation of growth factors in fibrin matrices. In this study, we present a slow release system for VEGF(165) and bFGF in fibrin sealant.

METHODS

In vitro: Pieces of Integratrade mark matrix of 15 mm in diameter were prepared. Integratrade mark matrices were divided into four groups (A=control; B=fibrin sealant; C=fibrin sealant+growth factors; D=growth factors). In vivo: The bioartificial dermal templates were transplanted into a full-skin defect of the back of nu-nu mice. Four different groups included each six matrices at 2 and 4 weeks.

RESULTS

In vitro: In groups C and D, continuous release of VEGF(165) and bFGF was eminent. The incorporation of growth factors into fibrin sealant evoked a prolonged growth factor release (p < 0.05). In vivo: A significantly higher amount of vessels was quantified in groups C and D compared to groups A and B (p < 0.001).

CONCLUSIONS

A model of slow protein release by combining VEGF(165) and bFGF with fibrin sealant was produced. This model resulted in a prolonged bioavailability of growth factors in vivo for functional purposes. Fibrin and collagen can release growth factors in vivo and induce significant and faster neovascularisation in bioartificial dermal templates.

Injury to rat brain induces a 3-10-fold increase in the activity of <em>factors</em> capable of stimulating astrocyte DNA synthesis and cell division in vitro. Maximum mitogenic activity was reached 10-<em>15</em> days post-lesion in both the tissue surrounding the wound and in the gelfoam filling the wound cavity. <em>Factors</em> capable of transforming the astrocyte morphology from polygonal-flat to fibrous-like (morphogens) could also be observed in brain tissue and showed increased activity beginning at 10 days postlesion. On the other hand, morphogenic activity was very low or absent in gelfoam extracts until <em>15</em> days postlesion. Both mitogenic and morphogenic <em>factors</em> were nondiffusible and were partly temperature and trypsin sensitive, i.e. they had the properties of protein-like substances, but seemed different from both epidermal and <em>fibroblast</em> <em>growth</em> <em>factors</em>. As judged by their filtration behavior on Amicon membranes, the molecular weight of mitogens and morphogens ranged from lower than 30,000 to greater than 100,000. Inhibitors of both mitogenic and morphogenic activities with molecular weight lower than 30,000 seemed to be also present in the brain extracts. The <em>factors</em> described here can account for the processes of astrocytosis and astrogliosis observed in vivo in response to CNS injury.

We investigated whether stem cells (MDSC) from primary cultures of rat skeletal muscle can differentiate into the smooth muscle lineage in response to vascular endothelial <em>growth</em> <em>factor</em> (VEGF) and coculture with bladder smooth muscle cells. The MDSC were isolated from gastrocnemius muscle biopsies of normal 3-6 week-old Sprague-Dawley rats and purified by the preplate technique. Cells that took approximately 6 days to adhere to the collagen-coated flasks were termed late preplate (LP) cells, and were used in all the experiments. The early plate (EP) cells (pp1-pp4) contained some myogenic cells but were mostly <em>fibroblasts</em> (< <em>15</em>% desmin+ cells) whereas the LP cells (pp5-pp6) were highly purified muscle-derived cells (pp6) >> 90% desmin+ cells). The muscle-derived stem cells (LP cells) were CD34+ or Sca-1+, CD45- and desmin+ by immunohistochemical staining. After two days of co-culture with bladder smooth muscle cells, about 25% of the muscle-derived stem cells were positive for alpha-smooth muscle actin (alpha-SMA)+. RT-PCR for alpha-SMA was positive in the VEGF stimulated MDSC, but negative in the absence of VEGF. In conclusion, rat muscle-derived stem cells exhibited stem cell properties (CD34+ or Sca-1+), and were not of hematogeous (CD45-) but of myogenic origin (desmin+). RT-PCR of alpha-SMA was positive in the VEGF stimulated muscle-derived stem cells.

We found aberrant DNA methylation of the WNT10B promoter region in 46% of primary hepatocellular carcinoma (HCC) and <em>15</em>% of colon cancer samples. Three of 10 HCC and one of two colon cancer cell lines demonstrated low or no expression, and 5-aza-2'deoxycytidine reactivated WNT10B expression with the induction of demethylation, indicating that WNT10B is silenced by DNA methylation in some cancers, whereas WNT10B expression is up-regulated in seven of the 10 HCC cell lines and a colon cancer cell line. These results indicate that WNT10B can be deregulated by either overexpression or silencing in cancer. We found that WNT10B up-regulated beta-catenin/Tcf activity. However, WNT10B-overexpressing cells demonstrated a reduced <em>growth</em> rate and anchorage-independent <em>growth</em> that is independent of the beta-catenin/Tcf activation, because mutant beta-catenin-transduced cells did not suppress <em>growth</em>, and dominant-negative hTcf-4 failed to alleviate the <em>growth</em> suppression by WNT10B. Although WNT10B expression alone inhibits cell <em>growth</em>, it acts synergistically with the <em>fibroblast</em> <em>growth</em> <em>factor</em> (FGF) to stimulate cell <em>growth</em>. WNT10B is bifunctional, one function of which is involved in beta-catenin/Tcf activation, and the other function is related to the down-regulation of cell <em>growth</em> through a different mechanism. We suggest that FGF switches WNT10B from a negative to a positive cell <em>growth</em> regulator.

The inner ear develops from a patch of thickened cranial ectoderm adjacent to the hindbrain called the otic placode. Studies in a number of vertebrate species suggest that the initial steps in induction of the otic placode are regulated by members of the <em>Fibroblast</em> <em>Growth</em> <em>Factor</em> (FGF) family, and that inhibition of FGF signaling can prevent otic placode formation. To better understand the genetic pathways activated by FGF signaling during otic placode induction, we performed microarray experiments to estimate the proportion of chicken otic placode genes that can be up-regulated by the FGF pathway in a simple culture model of otic placode induction. Surprisingly, we find that FGF is only sufficient to induce about <em>15</em>% of chick otic placode-specific genes in our experimental system. However, pharmacological blockade of the FGF pathway in cultured chick embryos showed that although FGF signaling was not sufficient to induce the majority of otic placode-specific genes, it was still necessary for their expression in vivo. These inhibitor experiments further suggest that the early steps in otic placode induction regulated by FGF signaling occur through the MAP kinase pathway. Although our work suggests that FGF signaling is necessary for otic placode induction, it demonstrates that other unidentified signaling pathways are required to co-operate with FGF signaling to induce the full otic placode program.

The effects of insulin, somatomedin-C (Sm-C), epidermal <em>growth</em> <em>factor</em> (EGF), <em>fibroblast</em> <em>growth</em> <em>factor</em> (FGF), vitamin E, and retinoic acid on <em>growth</em> and function of immature cultured pig Sertoli cells were investigated. All these <em>factors</em>, except vitamin E, stimulated Sertoli cell DNA synthesis and proliferation. The mitogenic effects of insulin observed only at micromolar concentrations were similar to those induced by nanomolar concentrations of Sm-C or EGF, but significantly less than those induced by FGF. The effects of EGF and Sm-C were almost additive, whereas those of Sm-C and FGF were synergistic. After a 6-day treatment, FGF and retinoic acid induced a significant increase in the number of follicle-stimulating hormone (FSH) receptors per cell, and in FSH-induced cyclic adenosine 3',5'-monophosphate (cAMP) production. Sm-C, which alone had no effect on these two parameters, potentiated FGF action. Basal plasminogen activator activity was enhanced after the 6-day treatment with EGF plus insulin and, particularly, with FGF plus insulin. Similarly, the response of the latter group to FSH was significantly higher than in any other group of cells. FGF was also able to stimulate cell multiplication and enhanced the FSH receptor number of Sertoli cells isolated from <em>15</em>- and 26-day-old rats. Thus, FGF is the most potent known mitogenic <em>factor</em> for cultured Sertoli cells, and it stimulates the phenotypic expression of these cells.

Transforming <em>growth</em> <em>factor</em>-beta (TGF-beta) is a ubiquitous <em>growth</em>-regulating protein that is capable of influencing the <em>growth</em> and function of heart cells in vitro. To better understand the role TGF-beta might play as a paracrine mediator of cardiac hypertrophy, the expression, secretion, and <em>growth</em> effects of TGF-beta were examined. Neonatal cardiac <em>fibroblasts</em> in vitro secreted latent TGF-beta 1 and TGF-beta 2 as high as <em>15</em> ng/10(6) cells. Angiotensin II (ANG II) and norepinephrine (NE) each augmented up to threefold the expression and secretion of latent TGF-beta 1 and TGF-beta 2 and also induced a shift in isoform predominance from beta 1 to beta 2. Each agent individually produced hypertrophic <em>growth</em> of neonatal cardiocytes and hyperplastic <em>growth</em> of cardiac <em>fibroblasts</em>. Paradoxically, the combination of NE and ANG II at intermediate and high concentrations resulted in less TGF-beta secretion (compared with either agent alone) and in hypertrophic <em>growth</em> of <em>fibroblasts</em>. These results suggest that the <em>growth</em>-promoting effects of ANG II and NE may in part be mediated via a paracrine stimulation of TGF-beta secretion.

Previous studies in this laboratory have indicated that the early embryonic chick heart depends on <em>fibroblast</em> <em>growth</em> <em>factor</em>-2 (FGF-2; bFGF), sequentially utilized in paracrine and autocrine fashion, for its <em>growth</em> and development (Sugi and Lough, [1995] Dev. Biol. 168-567-574). This view emanated from immunohistochemical detection of FGF-like antigens in endoderm cells at stage 6, and later in the early myocardium at stage 9+ (Parlow et al. [1991] Dev. Biol. 146:139-147). To identify other members of the FGF family that are expressed by these cells, we have used peptide-generated antisera that specifically recognize FGFs 1 and 4. Like FGF-2, FGFs 1 and 4 were exclusively detected in the endoderm at stage 5+ and later in the myocardium, appearing as punctate cytoplasmic deposits. However, whereas FGF-2 is first detected at stage 9+, FGFs 1 and 4 did not appear until stages 11 and <em>15</em>, respectively. Expression of all FGFs peaked at stages 18-24, decreasing thereafter in parallel with reduced myocardial cell proliferation. To determine these isoproteins' ability to facilitate the completion of terminal cardiac myocyte differentiation, stage 5+ precardiac mesoderm was cultured in defined medium with purified FGFs. Like FGF-2, as little as 5-10 ng/ml FGF-1 or FGF-4 supported the proliferation and differentiation of precardiac myoblasts, resulting in the formation of a vesicle containing an adherent multilayer of synchronously contractile cells. Evidence that this represented FGF receptor-mediated signaling rather than a nonspecific effect of exogenous FGF was indicated by the ability of sodium chlorate to inhibit FGF-mediated cardiogenesis. These findings are consistent with the hypothesis that, like FGF-2, FGFs 1 and 4 participate in the regulation of early heart development via paracrine and autocrine mechanisms.

Angiotensin II (ANG II), a potent vasoconstricting peptide, may act as a <em>growth</em> <em>factor</em> for cardiac muscle cells and induce hypertrophy. We examined the molecular phenotype of neonatal rat cardiac <em>fibroblasts</em> in relation to ANG II by studying the expression pattern of three transcription <em>factors</em> (Egr-1, c-fos and c-jun) and the transforming <em>growth</em> <em>factor</em>-beta 1 (TGF-beta 1). ANG II did not affect cell proliferation and <em>growth</em> of serum deprived neonatal cardiac <em>fibroblasts</em> as predicted from their DNA and protein contents. The expression of Egr-1 and c-fos was induced as early as <em>15</em> min that reached maximal levels at 45 min and declined thereafter, whereas c-jun was induced at 45 min and remained elevated up to 2 hrs of ANG II addition. ANG II up-regulated the expression of TGF-beta 1, which became apparent after 1 hr of incubation and reached a plateau between 16-48 hrs. Our results indicate that ANG II transiently stimulates the expression of transcription <em>factors</em>, which may up-regulate TGF-beta 1, that in turn could contribute to the process of myocardial extra-cellular matrix remodeling in hypertrophy.

OBJECTIVE

Malignant gliomas are typically angiogenic and express greater amounts of angiogenic factors. We examined glioma tissues for their expression of an endogenous inhibitor of angiogenesis, endostatin, a COOH-terminal fragment of collagen XVIII.

METHODS

We examined frozen tissues from 51 patients with astrocytic tumors (grade 2, 13; grade 3, 9; and grade 4, 29). Frozen tissues were subjected to immunoblot analysis and immunohistochemistry for endostatin. Tumor vascular density was determined by calculating the percentage of tumor capillary vessel areas/tissue section area. Tissue concentrations of vascular endothelial growth factor and basic fibroblast growth factor were examined by enzyme immunoassay.

RESULTS

The levels of endostatin protein estimated by immunoblotting were significantly higher in grade 4 than lower-grade glioma tissues. The immunoreactive bands for endostatin were identified as the fragment derived from noncollagenous domain 1 of collagen XVIII, a peptide 15 residues longer than endostatin toward the NH(2)-terminal end, by NH(2)-terminal amino acid sequencing. In addition to an intense immunoreactivity for endostatin in tumor blood vessels, sections from malignant gliomas showed widely distributed immunoreactivity around tumor cells near the hyperplastic microvessels. The tumor vascular density and the levels of vascular endothelial growth factor in grade 4 glioma tissues were significantly higher than grade 2 and grade 3 gliomas, whereas the levels of basic fibroblast growth factor were the same.

CONCLUSIONS

The results indicate a positive correlation between the levels of tissue endostatin and malignancy grades in gliomas. The endostatin may be released near the tumor blood vessels with hyperplasia to counteract angiogenic stimuli in malignant gliomas.

Placental blood flow, nitric-oxide (NO) levels, and endothelial NO synthase (eNOS) expression increase during human and ovine pregnancy. Shear stress stimulates NO production and eNOS expression in ovine fetoplacental artery endothelial (OFPAE) cells. Because eNOS is the rate-limiting enzyme essential for NO synthesis, its activity and expression are both closely regulated. We investigated signaling mechanisms underlying pulsatile shear stress-induced increases in eNOS phosphorylation and protein expression by OFPAE cells. The OFPAE cells were cultured at 3 dynes/cm2 shear stress, then exposed to <em>15</em> dynes/cm2 shear stress. Western blot analysis for phosphorylated ERK1/2, Akt, p38 mitogen activated protein kinase (MAPK), and eNOS showed that shear stress rapidly increased phosphorylation of ERK1/2 and Akt but not of p38 MAPK. Phosphorylation of eNOS Ser1177 under shear stress was elevated by 20 min, a response that was blocked by the phosphatidyl inositol-3-kinase (PI-3K)-inhibitors wortmannin and LY294002 but not by the mitogen activated protein kinase kinase (MEK)-inhibitor UO126. Basic <em>fibroblast</em> <em>growth</em> <em>factor</em> (bFGF) enhanced eNOS protein levels in static culture via a MEK-mediated mechanism, but it could not further augment the elevated eNOS protein levels otherwise induced by the <em>15</em> dynes/cm2 shear stress. Blockade of either signaling pathway changed the shear stress-induced increase in eNOS protein levels. In conclusion, shear stress induced rapid eNOS phosphorylation on Ser1177 in OFPAE cells through a PI-3K-dependent pathway. The bFGF-induced rise in eNOS protein levels in static culture was much less than those observed under flow and was blocked by inhibition of MEK. Prolonged shear stress-stimulated increases in eNOS protein were not affected by inhibition of MEK- or PI-3K-mediated pathways.

The small GTPase RhoA plays a critical role in signaling pathways activated by serum-derived <em>factors</em>, such as lysophosphatidic acid (LPA), including the formation of stress fibers in <em>fibroblasts</em> and neurite retraction and rounding of soma in neuronal cells. Previously, we have shown that ectopic expression of v-Crk, an SH2/SH3 domain-containing adapter proteins, in PC12 cells potentiates nerve <em>growth</em> <em>factor</em> (NGF)-induced neurite out<em>growth</em> and promotes the survival of cells when NGF is withdrawn. In the present study we show that, when cultured in <em>15</em>% serum or lysophosphatidic acid-containing medium, the majority of v-Crk-expressing PC12 cells (v-CrkPC12 cells) display a flattened phenotype with broad lamellipodia and are refractory to NGF-induced neurite out<em>growth</em> unless serum is withdrawn. v-Crk-mediated cell flattening is inhibited by treatment of cells with C3 toxin or by mutation in the Crk SH2 or SH3 domain. Transient cotransfection of 293T cells with expression plasmids for p160ROCK (Rho-associated coiled-coil-containing kinase) and v-Crk, but not SH2 or SH3 mutants of v-Crk, results in hyperactivation of p160ROCK. Moreover, the level of phosphatidylinositol-4,5-bisphosphate is increased in v-CrkPC12 cells compared to the levels in mutant v-Crk-expressing cells or wild-type cells, consistent with PI(4)P5 kinase being a downstream target for Rho. Expression of v-Crk in PC12 cells does not result in activation of Rac- or Cdc42-dependent kinases PAK and S6 kinase, demonstrating specificity for Rho. In contrast to native PC12 cells, in which focal adhesions and actin stress fibers are not observed, immunohistochemical analysis of v-CrkPC12 cells reveals focal adhesion complexes which are formed at the periphery of the cell and are connected to actin cables. The formation of focal adhesions correlates with a concomitant upregulation in the expression of focal adhesion proteins FAK, paxillin, alpha3-integrin, and a higher-molecular-weight form of beta1-integrin. Our results indicate that v-Crk activates the Rho-signaling pathway and serves as a scaffolding protein during the assembly of focal adhesions in PC12 cells.

Bilayered living human skin equivalent (HSE) consists of cultured keratinocytes residing on the surface of a fibroblast-populated collagen lattice. Although HSE is FDA-approved for treatment of diabetic foot and venous stasis ulcers, its clinical efficacy remains limited, because the molecular mechanisms underlying its therapeutic effect are not fully understood. It is, therefore, often applied mistakenly as a skin graft. In this report, we delineate a mechanism of HSE biological effect and consequent optimal clinical use in accelerating closure of diabetic foot ulcers.

METHODS

HSE was grafted onto nude mice and the release of various growth factors was evaluated by reverse transcription-polymerase chain reaction (RT-PCR) and immunochemistry. Clinical: HSE was grafted onto 11 consecutive patients with diabetes who had 13 non-ischemic foot ulcers and healing was measured as time to 100% closure (e.g., no drainage and 100% epithelialized).

METHODS

HSE cellular components were determined to express 15 different growth factors/cytokine genes known to promote wound healing. Histological evidence from the nude mice showed that the collagen component of HSE underwent remodeling within the first seven days of grafting. Clinical: All diabetic foot ulcers healed in 31.8 12.4 days. Local release of a unique combination of 15 growth factors expressed by HSE keratinocyte and fibroblast components generates closure of diabetic foot ulcers. HSE should be applied with the same surgical conditions for a skin graft (i.e., no cellulitis, no drainage, and negligible bacteria). We hypothesize that bilayered HSE generates its effect by way of the local synthesis and release of multiple growth factors in specific combination and concentration, which improves the impaired reparative process of chronic wounds.

The purpose of this study was to characterize the pattern of cytokine gene expression by human corneal epithelial cells (HCEC) in response to interleukin-1 (IL-1). Primary cultured HCEC (P-HCEC) or SV40 transformed HCEC (SV40-HCEC) were treated for 6 hr with serum-free <em>growth</em>-media alone or with recombinant human IL-1beta or IL-1alpha (10 ng ml(-1)). 33P labeled cDNA was generated from total RNA, then hybridized to a human cytokine expression array. An autoradiograph was generated for each experimental condition and results analysed semi-quantitatively. Reverse transcription polymerase chain reaction (RT-PCR) was performed to detect mRNA for IL-8, <em>growth</em> related oncogene-beta (GRO-beta), intercellular adhesion molecule (ICAM)-1 and Ephrin A5. P-HCEC and SV40-HCEC demonstrated comparable cytokine profiles. For P-HCEC (n=2) the expression of 35 genes was upregulated or only detectable following IL-1beta treatment whereas the expression of nine genes was downregulated or undetectable after IL-1beta treatment. In SV40-HCEC (n=3), the expression of 48 genes was upregulated or only detectable following IL-1beta treatment and the expression of 10 genes was downregulated or undetectable after IL-1beta treatment. Some genes that demonstrated increased expression included cadherin-5, ICAM-1, GRO-alpha, GRO-beta, GRO-gamma, Activin A (bA subunit), tumor necrosis <em>factor</em>-alpha, IL-6, and IL-8. Genes that showed decreased expression included the chemokine receptor-CXCR-4, ciliary neurotrophic <em>factor</em> (CNTF), c-kit ligand, Ephrin A5, G-protein coupled receptor RDC-1 and FGF family FGFR2. Bayesian analysis of the SV40-HCEC data (n=3) revealed the expression of <em>15</em> genes that were significantly (p<0.05) differentially regulated. Within these <em>15</em> genes, the expression of chemokines (GRO-alpha, GRO-beta, IL-8), <em>fibroblast</em> <em>growth</em> <em>factor</em> 13 and the cytokine IL-6 were the most upregulated, while ephrin A5 and chemokine receptor-4 were the most downregulated. IL-1alpha treatment (n=1 P-HCEC; n=1 SV40-HCEC) produced results very similar to IL-1beta treatment. RT-PCR revealed differential regulation of IL-8, GRO-beta, ICAM-1 and ephrin A5 in accordance with gene array data. In conclusion, the data demonstrate that IL-1 treatment of HCEC differentially regulates the expression of other cytokine and related genes, thus adding to the body of evidence that IL-1 is a major mediator of ocular surface inflammatory reactions. Since the expression of a large number of genes can be studied simultaneously, gene array studies such as these offers the advantage of understanding global changes in response to a specific stimulus. Thus our study provides insight in to the ocular surface response in conditions of inflammation and corneal wound healing where the levels of IL-1 are known to be increased.

The molecular events responsible for tumor progression in human cutaneous malignant melanoma remain unclear; however, critical to the process is the dysregulated proliferation of tumor cells and the development of new vascular channels which allow further <em>growth</em> and dissemination. Connective tissue mast cells (MC) have been implicated in tumor progression because they concentrate around tumors (including melanomas) prior to the formation of new blood vessels, and because they contain many chemical mediators, including basic <em>fibroblast</em> <em>growth</em> <em>factor</em> (bFGF), known to have mitogenic and angiogenic effects. Several MC chemotactic and mitogenic <em>factors</em> have been described including interleukin-3 (IL-3). In order to determine whether there is a differential expression of this MC chemotactic/mitogenic <em>factor</em> with tumor progression in vivo, we evaluated by immunohistochemistry 85 melanocytic lesions including primary invasive malignant melanoma (PIMM), melanoma in situ (MMIS), and ordinary intradermal benign melanocytic nevi (BMN) for expression of IL-3. Nucleic acid in situ hybridization also was used to evaluate the melanocytic lesions for IL-3-specific mRNA transcripts. Intracellular IL-3 protein was detected in 29/33 (88%) PIMM and <em>15</em>/25 (60%) MMIS, but was not detected in any (0/27; 0%) BMN (p < 0.0001). IL-3-specific mRNA transcripts were present in 3/4 PIMM and 2/10 MMIS in which IL-3 protein was not identified, but were not detected in any BMN. IL-3 mRNA or protein was not detected in normal melanocytes present in the perilesional epidermis of any of the specimens studied. Immunohistochemistry also was used to confirm the presence of IL-3 alpha-specific receptors on human cutaneous MC. As demonstrated by others, a significantly increased number of MC was present in the perilesional stroma of PIMM and MMIS vis a vis BMN (p < 0.0001). The results suggest that melanoma cells may attract MC in vivo by producing MC chemotactic/mitogenic <em>factors</em> such as IL-3. The recruitment of MC and the subsequent release of their potent mitogenic and angiogenic <em>factors</em> such as bFGF may thus represent a tumor-host interaction which favors tumor progression. Reed JA, McNutt NS, Bogdany JK, Albino AP. Expression of the mast cell <em>growth</em> <em>factor</em> interleukin-3 in melanocytic lesions correlates with an increased number of mast cells in the perilesional stroma: implications for melanoma progression.

A receptor for acidic and basic <em>fibroblast</em> <em>growth</em> <em>factors</em> (aFGF and bFGF, respectively) was isolated from 7-day embryonic chick. Chromatography of solubilized membrane proteins on wheat germ agglutininagarose and aFGF-Sepharose yielded three major polypeptides migrating at <em>15</em>0, 70, and 45 kDa as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. These polypeptides were eluted from aFGF-Sepharose with either 1.0 M NaCl or 100 micrograms/ml heparin, but were not retained on underivatized Sepharose. Cross-linking of 125I-aFGF or 125I-bFGF to either crude membrane preparations or to purified fractions yielded a 165-kDa complex, suggesting the existence of a <em>15</em>0-kDa FGF receptor after subtraction of approximately <em>15</em> kDa for 125I-FGF. Addition of excess aFGF or bFGF competed for binding of either 125I-aFGF or 125I-bFGF to FGF receptor preparations. Purified FGF receptor fractions were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to Immobilon membranes, and incubated with 125I-aFGF or 125I-bFGF in order to identify FGF-binding polypeptides. Bound 125I-aFGF and 125I-bFGF were displaced by aFGF and bFGF, but not epidermal <em>growth</em> <em>factor</em>, consistent with the identification of the <em>15</em>0-kDa polypeptide as a receptor for acidic and basic FGF. Treatment of purified FGF receptor fractions with N-glycanase demonstrated that the <em>15</em>0-kDa polypeptide contained approximately 10 kDa of N-linked oligosaccharide. The apparent molecular mass of the <em>15</em>0-kDa polypeptide was unaffected by treatment with heparitinase, indicating that the <em>15</em>0-kDa polypeptide is not a heparan sulfate proteoglycan. Together, these data suggest that the <em>15</em>0-kDa polypeptide is a FGF receptor that may mediate the biological activities of aFGF and bFGF.

The role of basic <em>fibroblast</em> <em>growth</em> <em>factor</em> (bFGF) and insulin-like <em>growth</em> <em>factor</em> I (IGF-I) in cartilage <em>growth</em> was studied in primary cultures of rat rib <em>growth</em> plate chondrocytes. <em>Growth</em> <em>factors</em> effects on expression of the proto-oncogene c-fos, DNA synthesis, differentiation, and morphological changes were analyzed by in situ hybridization, 3H-thymidine incorporation, and light and fluorescence microscopy. In serum-deprived cells, bFGF induced a transient expression of c-fos with a maximal effect <em>15</em>-30 minutes after stimulation. After 24 h of culture it had a slightly lower stimulatory effect on DNA synthesis than IGF-I, but became a significantly more potent mitogen than IGF-I after 48 and 72 h. The stimulatory effect of bFGF on DNA synthesis coincided with a decrease in collagen type II and IGF-II expression. In contrast, IGF-I alone stimulated expression of these genes. In bFGF-treated cultures, cell morphology and the appearance of actin filaments was changed. Polygonal chondrocytes became elongated, <em>fibroblast</em>-like, and the smooth actin filaments were brush-like and disrupted. Addition of IGF-I reduced these changes without affecting c-fos expression induced by bFGF. Our results suggest that bFGF stimulates cell proliferation by preventing terminal differentiation of chondrocytes. This effect is mediated by induction of c-fos expression and a decrease in the steady-state levels of transcripts for collagen II and IGF-II.

Xenopus oocytes expressing <em>fibroblast</em> <em>growth</em> <em>factor</em> receptor 1 (FGFR1) were used as a biological model system to analyse the signal transduction pathways that are triggered by <em>fibroblast</em> <em>growth</em> <em>factor</em> 1 (FGF1). Germinal vesicle breakdown (GVBD) and phosphorylation of extracellular signal-regulated protein kinase 2 (ERK2) occured <em>15</em> h after FGF1 addition. These events were Ras-dependent as they were blocked by a Ras dominant negative form. The Ras activity was promoted by three upstream effectors, <em>growth</em> <em>factor</em>-bound protein 2 (Grb2), phosphatidylinositol 3-kinase (PI3K) and Src cytoplasmic kinase. Ras activation was inhibited by a Grb2 dominant negative form (P49L), by PI3K inhibitors, including wortmannin, LY294002, the N-SH2 domain of p85alpha PI3K and by the SH2 domain of Src. Src activation induced by FGF1 was blocked by the SH2 domain of Src and PP2, a specific inhibitor of Src. The Grb2 adaptor was recruited by the upstream Src homology 2/alpha-collagen-related (Shc) effector, as the SH2-Shc domain prevented the GVBD and the ERK2 phosphorylation induced by FGF1. The importance of another signalling pathway involving phospholipase Cgamma (PLCgamma) was also investigated. The use of the PLCgamma inhibitory peptide, neomycin and the calcium chelator BAPTA-AM on oocytes expressing FGFR1 or the stimulation by PDGF-BB of oocytes expressing PDGFR-FGFR1 mutated on the PLCgamma binding site, prevented GVBD and ERK2 phosphorylation. This study shows that the transduction cascade induced by the FGFR1-FGF1 interaction in Xenopus oocytes represents the sum of Ras-dependent and PLCgamma-dependent pathways. It emphasizes the role played by PI3K and Src and their connections with the Ras cascade in the FGFR1 signal transduction.

Microsomal prostaglandin E synthase-1 (mPGES-1) is the terminal enzyme regulating the synthesis of prostaglandin E2 (PGE2) in inflammatory conditions. In this study we investigated the regulation of mPGES-1 in gingival <em>fibroblasts</em> stimulated with the inflammatory mediators interleukin-1 beta (IL-1beta) and tumour necrosis <em>factor</em> alpha (TNFalpha). The results showed that IL-1beta and TNFalpha induce the expression of mPGES-1 without inducing the expression of early <em>growth</em> response <em>factor</em>-1 (Egr-1). Treatment of the cells with the PLA2 inhibitor 4-bromophenacyl bromide (BPB) decreased the cytokine-induced mPGES-1 expression accompanied by decreased PGE2 production whereas the addition of arachidonic acid (AA) upregulated mPGES-1 expression and PGE2 production. The protein kinase C (PKC) activator PMA did not upregulate the expression of mPGES-1 in contrast to COX-2 expression and PGE2 production. In addition, inhibitors of PKC, tyrosine and p38 MAP kinase markedly decreased the cytokine-induced PGE2 production but not mPGES-1 expression. Moreover, the prostaglandin metabolites PGE2 and PGF2alpha induced mPGES-1 expression as well as upregulated the cytokine-induced mPGES-1 expression indicating positive feedback regulation of mPGES-1 by prostaglandin metabolites. The peroxisome proliferator-activated receptor-gamma (PPARgamma) ligand, <em>15</em>-deoxy-Delta12,14-prostaglandin J2 (<em>15</em>d-PGJ2), decreased mPGES-1 expression but not COX-2 expression or PGE2 production. The results indicate that the inflammatory-induced mPGES-1 expression is regulated by PLA2 and <em>15</em>d-PGJ2 but not by PKC, tyrosine kinase or p38 MAP kinase providing new insights into the regulation of mPGES-1.

Pigment epithelium-derived <em>factor</em> (PEDF) exerts anti-angiogenic actions. However, the signal-transduction pathways regulated by PEDF remain to be elucidated. We show here that PEDF inhibited <em>fibroblast</em> <em>growth</em> <em>factor</em> 2 (FGF-2) induced capillary morphogenesis of a murine brain capillary endothelial cell line (IBE cells) and of human umbilical-vein endothelial cells (HUVECs) cultured on <em>growth</em>-<em>factor</em>-reduced Matrigel. We previously showed that FGF-2-mediated capillary morphogenesis was blocked by the Src-kinase inhibitor PP2 and that expression of dominant negative Fyn in IBE cells inhibited capillary morphogenesis. We examined the effect of PEDF on kinase activity of Fyn and found that PEDF downregulated FGF-2-promoted Fyn activity by tyrosine phosphorylation at the C-terminus in a Fes-dependent manner. In a stable IBE cell line expressing kinase-inactive Fes (KE5-<em>15</em> Fes cells), PEDF failed to inhibit FGF-2-induced capillary morphogenesis or Fyn activity. PEDF induced the colocalization of Fyn and Fes in IBE cells expressing wild-type Fes, but not in KE5-<em>15</em> Fes cells. In addition, wild-type Fes increased the tyrosine phosphorylation of Fyn in vitro, suggesting that Fes might directly phosphorylate Fyn. Expression of constitutively active Fyn (Y531F) in IBE cells exhibited capillary morphogenesis in the absence of FGF-2 and was resistant for PEDF treatment. Our results suggest that PEDF downregulates Fyn through Fes, resulting in inhibition of FGF-2-induced capillary morphogenesis of endothelial cells.

BACKGROUND

The study was designed to compare the bone anabolic effects of basic fibroblast growth factor (bFGF), a selective agonist for prostaglandin E receptor subtype EP4, and parathyroid hormone (PTH) in aged ovariectomized (OVX) rats with severe cancellous osteopenia.

METHODS

Groups of aged OVX rats were maintained untreated for 1 year postovariectomy (15 months of age) to develop severe tibial cancellous osteopenia. These animals were then treated with bFGF or the EP4 agonist (EP4) for 3 weeks. Other groups of aged OVX rats were treated with EP4 or PTH alone for 11 weeks, or sequentially with bFGF or EP4 for 3 weeks followed by PTH for 8 weeks. Cancellous and cortical bone histomorphometry were performed in the right proximal tibial metaphysis and tibial diaphysis respectively.

RESULTS

Treatment with bFGF for 3 weeks markedly increased serum osteocalcin, osteoid volume, and osteoblast and osteoid surfaces to a greater extent than EP4. Basic FGF, but not EP4 or PTH, induced formation of osteoid islands within bone marrow. EP4 stimulated cancellous bone turnover, but failed to restore lost cancellous bone in the severely osteopenic proximal tibia after 11 weeks of treatment. In contrast, EP4, much like PTH, increased cortical bone mass in the tibial diaphysis by stimulating both periosteal and endocortical bone formation. Treatment of aged OVX rats with PTH alone tended to partially reverse the severe tibial cancellous osteopenia, whereas sequential treatment with bFGF and PTH increased tibial cancellous bone mass to near the level of vehicle-treated control rats. These findings indicate that bFGF had the strongest stimulatory effect on cancellous bone formation, and was the only anabolic agent to induce formation of osteoid islands within the bone marrow of the severely osteopenic proximal tibia. Therefore, bFGF may be more effective for the reversal of severe cancellous osteopenia. PTH and EP4 increased cortical bone mass to nearly the same extent, but cancellous bone mass was greater by two-fold in PTH-treated OVX rats than in EP4-treated OVX rats.

CONCLUSIONS

These findings in aged OVX rats suggest that PTH is more efficacious than EP4 for augmentation of cancellous bone in the severely osteopenic, estrogen-deplete skeleton.

We investigated the binding properties of the type I insulin-like <em>growth</em> <em>factor</em> (IGF) receptor expressed in NIH-3T3 <em>fibroblasts</em> transfected with a human type I receptor cDNA. Cell surface receptors bound IGF-I with KD = 1 nM as predicted. Although recent studies have suggested that IGF-I and IGF-II bind to type I receptors with near-equal affinity, the receptors in this system bound IGF-II with much lower affinity (KD = <em>15</em>-20 nM). When type I receptors from the transfected cells were solubilized and immunopurified, however, both 125I-IGF-I and 125I-IGF-II bound to the purified receptors with extremely high and relatively similar affinities (KD = 8 and 17 pM respectively). Thus the immunopurified receptors had higher affinity but lower specificity for the two ligands. The monoclonal antibody alpha IR-3 effectively inhibited IGF-I binding to cell surface receptors (75 +/- 10%), but did not inhibit IGF-II binding. In the purified receptor assay, alpha IR-3 also inhibited IGF-I binding more effectively than IGF-II binding (38 +/- 7% versus 10 +/- 4%). We conclude that the products of this cDNA can account for the binding patterns that we previously observed in receptors immunopurified from human placenta. The differential effect of alpha IR-3 on IGF-I versus IGF-II raises the possibility that these homologous <em>growth</em> <em>factors</em> bind to immunologically distinct epitopes on the type I receptor.

Cardiac tissue engineering presents a challenge due to the complexity of the muscle tissue and the need for multiple signals to induce tissue regeneration in vitro. We investigated the effects of compression (1 Hz, <em>15</em>% strain) combined with fluid shear stress (10(-2) -10(-1) dynes/cm(2) ) provided by medium perfusion on the outcome of cardiac tissue engineering. Neonatal rat cardiac cells were seeded in Arginine-Glycine-Aspartate (RGD)-attached alginate scaffolds, and the constructs were cultivated in a compression bioreactor. A daily, short-term (30 min) compression (i.e., "intermittent compression") for 4 days induced the formation of cardiac tissue with typical striation, while in the continuously compressed constructs (i.e., "continuous compression"), the cells remained spherical. By Western blot, on day 4 the expression of the gap junction protein connexin 43 was significantly greater in the "intermittent compression" constructs and the cardiomyocyte markers (α-actinin and N-cadherin) showed a trend of better preservation compared to the noncompressed constructs. This regime of compression had no effect on the proliferation of nonmyocyte cells, which maintained low expression level of proliferating cell nuclear antigen. Elevated secretion levels of basic <em>fibroblast</em> <em>growth</em> <em>factor</em> and transforming <em>growth</em> <em>factor</em>-β in the daily, intermittently compressed constructs likely attributed to tissue formation. Our study thus establishes the formation of an improved cardiac tissue in vitro, when induced by combined mechanical signals of compression and fluid shear stress provided by perfusion.

Although in the normal healthy organism angiogenesis is a tightly regulated process, under a variety of circumstances it may contribute to disease states. These include the <em>growth</em> of solid tumors, the hematogenous spread of tumor cells and the <em>growth</em> of metastasis. Our aim was to measure the levels of 5 angiogenic cytokines in the plasma of patients with a variety of cancers, to establish a plasmatic angiogenic profile. We prospectively obtained blood samples in citrated tubes from 40 healthy individuals and 75 patients with a variety of solid tumors. Patients who had received any form of treatment in the preceeding 6 months were excluded from the study. Plasma levels of the following 5 cytokines were determined by ELISA: vascular endothelial <em>growth</em> <em>factor</em> (VEGF), hepatocyte <em>growth</em> <em>factor</em> (HGF), basic <em>fibroblast</em> <em>growth</em> <em>factor</em>, transforming <em>growth</em> <em>factor</em>-beta and tumor necrosis <em>factor</em>-alpha. In some cases, additional samples were taken 4 and <em>15</em> days after surgical removal of the tumor. Our findings demonstrate, that firstly, compared to the tumor group VEGF was almost always undetectable or present at very low levels in healthy individuals; secondly, a threshold value for HGF was found to exist between the 2 groups (healthy vs. tumor); and thirdly, there was a clear relationship between plasma levels of VEGF and HGF and extension of disease (i.e., without or with metastases). The timing of blood sampling in the post-operative period was found to be critical, particularly with regard to VEGF and HGF. The existence of a systemic angiogenic profile in the plasma of cancer patients may be useful as a diagnostic and prognostic tool and may help in the future to monitor the responses of individual patients to anti-tumor and, particularly, anti-angiogenic therapy.